In Section 3-6, you began to learn about experimental studies. In this section, I will continue this discussion in order to help you become familiar with some other aspects of experimental studies.

To see why we can make cause-and-effect inferences from the results of experimental studies, let's look again at the results of a fictional correlational study measuring, in a sample of introductory-psychology students, two variables: whether or not caffeine was consumed while studying for a test (obtained from soda, coffee, tea, etc.) and test scores. Let's say that, in this correlational study, the researchers found the same fictional results presented in Figure 1 of the previous section (that is, an average score of about 63% for those who consumed caffeine and an average score of about 53% for thise who didn't). In this case, therefore, there was a positive correlation between the reported use of caffeine and test scores. Could we then conclude from the results of this correlational study that the use of caffeine while studying causes higher test scores, on average?

No, we couldn't. You may have noticed that there are several features of the correlational study that make questionable any attempt to conclude that using caffeine causes higher test scores. Some of these problematic features are:

• Self-report data are limited by memory distortions, lack of awareness (some may not have realized that chocolate contains caffeine), and lying.
• Caffeine intake occurred through various means (soda, coffee, tea, food, caffeine pills, etc.), which may change the effects of this drug.
• Different amounts of caffeine probably were used by different participants.
• It may be that participants who were more interested in the course studied more and ingested more caffeine to stay alert while studying.

In other words, correlational studies do not control for extraneous variables and the directionality of causation. Because of this, there are many possible causal explanations for the results of a correlational study.

To review: In experimental studies, there is an attempt to exclude all alternative explanations for the results but the one being investigated. By manipulating the causal variable directly, researchers make sure that this variable comes first, thereby resolving the directionality problem. By using a method (such as random assignment) that increases the likelihood that the distribution of important extraneous variables is similar in the two groups, researchers make sure that only changes in the causal variable can explain changes in the affected variable, thereby resolving the third-variable problem and controlling for the effects of other extraneous variables.

Let's look at one more example of an experimental study in order to introduce the correct terminology for the groups and variables in these studies. Triplett (1898) observed that bicycle racers have faster times when they race against others than when they race alone. Based on this observation, he hypothesized that the presence of competitors causes people to perform tasks faster. With respect to bicycle racing, Triplett's hypothesis stated that:

the bodily presence of another rider is a stimulus to the racer in arousing the competitive instinct; that another can thus be the means of releasing or freeing nervous energy for him that he cannot of himself release; and, further, that the sight of movement in that other by perhaps suggesting a higher rate of speed, is also an inspiration to greater effort. (p. 516)

In order to test this hypothesis, Triplett (1898) designed an experimental study in which he asked children (from about 9 to 13 years of age) to wind fishing line onto a fishing reel as fast as possible. The children performed the task under two conditions: either alone or in the presence of a competitor. The children sat at the fishing reel labelled "a" on the left side of Figure 1, and their speed was measured by a line traced onto a piece of paper attached to a rotating drum (called a kymograph) at the other end of the apparatus. A competitor sat at the reel labelled "B" on the left side of that figure.

Figure 1. The Y-shaped apparatus used in the experimental study performed by Triplett (1898)

Triplett's (1898) "causal variable" was peforming the task alone or with a competitor. In an experiment, the causal variable is referred to as the independent variable, which is defined as the variable hypothesized to cause changes in a second variable. The independent variable is directly manipulated by the experimenters.

Triplett's "affected variable" was the speed with which the fishing line is wound onto the reel. In an experiment, the affected variable is referred to as the dependent variable, which is defined as the variable hypothesized to be affected by changes in the independent variable.

Random assignment was not necessary in Triplett's study because each child was subjected to both levels of the independent variable — that is, each child performed the task alone and then, at another time, they performed the task with another child present. Extraneous variables, therefore, could not explain an average difference between the two conditions because each child brought the same extraneous variables into each condition. Triplett's experimental design is summarized in Table 1:

Table 1. The design of Triplett's (1898) study of "social facilitation"

The two conditions are labelled "experimental" and "control." The experimental condition refers to the situation or event hypothesized to cause a particular change in the dependent variable. This change often is a desired one, so that one might refer to the experimental condition as the treatment condition. Because Triplett hypothesized that the presence of a competitor increases the speed of task performance, the situation in which a competitor was present was the experimental condition. In experimental studies, the participants subjected to the experimental condition make up the experimental group.

In order to determine whether or not the experimental condition has affected the dependent variable in the way hypothesized, participants are subjected to a second condition called the control condition, which does not include the situation or event hypothesized to cause a particular change in the dependent variable. The specifics of the control condition depend upon the hypothesis being tested and the extraneous factors thought to affect the dependent variable. Devising an adequate control condition is perhaps the most difficult aspect of experimental design. In experimental studies, the participants subjected to the control condition make up the control group.

As he had predicted, Triplett (1898) found that most (but not all) children wound the fishing line more rapidly in the presence of a competitor. The general phenomenon demonstrated by Triplett is called social facilitation, which refers to the faster (or better) performance of a task when others are present. Triplett's study was one of the first to demonstrate this claim experimentally. In Section 7, you will learn more about the effects of social situations on our cognitions, emotions, and behavior.

Study Questions for Section 3-7

1. How is the experimental group similar to and different from the control group?
2. I wish to test the following hypothesis: using humor when teaching a class causes students to pay more attention to the class discussion. What are the independent and dependent variables in this example?
3. How would you design an experimental study to test the hypothesis that increasing dopamine activity in the brain (dopamine is an important brain chemical — a neurotransmitter) causes the development of psychotic symptoms in normal people (psychotic symptoms are things like hallucinations and delusions)? (NOTE: in your answer, please describe the independent variable, the dependent variable, the experimental group, and the control group.)
4. What was the hypothesis that Triplett (1898) was trying to test?
5. How could you use Triplett's results in your everyday life (for example, when performing a task in your job or at school)?

Go to Quiz 3-7 questions

Go to Readings Section 3-8